Evidence That the Mammary Fat Pad Mediates the Action of Growth Hormone in Mammary Gland Development

0013-7227/98/$03.00/0 Endocrinology Copyright © 1998 by The Endocrine Society Vol. 139, No. 2 Printed in U.S.A. Evidence That the Mammary Fat Pad Mediates the Action of Growth Hormone in Mammary Gland Development PAUL D. WALDEN, WEIFENG RUAN, MARK FELDMAN, AND DAVID L. KLEINBERG Departments of Medicine, Urology, and Biochemistry, Department of Veterans Affairs Medical Center and New York University Medical Center, New York, New York 10010 ABSTRACT Recent evidence from our laboratory suggests that GH and insulinlike growth factor I (IGF-I) mediate glandular mammary development together with estrogen. It has also been well established that both stromal and epithelial elements must interact for mammary glandular development to occur. To determine whether the effect of GH is mediated by the stromal or epithelial tissue, we set up the following experiment. Bovine GH (bGH; 100 mg) or BSA (as a control), without or with estradiol (E2), was injected ip into sexually immature female rats that were hypophysectomized and oophorectomized. Mammary glands and subscapular fat pads were removed from the animals. The mammary glands were divided into two parts: a glandfree fat pad and remaining glandular tissue. The end point of bGH activity was induction of IGF-I messenger RNA (mRNA). This was determined quantitatively by solution hybridization and also by RT- PCR. We found that the effects of GH on stimulation of IGF-I mRNA in the gland-free mammary fat pad and the remainder of the mammary gland were similar (3.6- vs. 3.9-fold, respectively; P , 0.001). In both sorts of mammary tissue, bGH was found to synergize with E2 in the induction of IGF-I mRNA (5.8- vs. 5.3-fold; P , 0.001). There was also an increase in IGF-I mRNA in subscapular fat pads in response to 100 mg bGH (5.3-fold; P , 0.001); however, no synergism between bGH and E2 was found. These data indicate that bGH works as well on mammary stromal tissue as on tissue with glands and suggests that GH acts on the stromal compartment of the mammary gland to induce IGF-I mRNA and possibly IGF-I itself, which, in turn, causes differentiation of epithelial ducts into terminal end buds. These data also might explain why mammary epithelium is also able to differentiate in nonmammary fat pads when transplanted there. (Endocrinology 139: 659 – 662, 1998) P UBERTAL mammary development occurs in response to an increase in estradiol (E2). However, mammary development cannot take place in the absence of the pituitary gland. The pituitary hormone necessary for mammary development is GH (1–5). In recent years, we found that GH acts on the mammary gland through specific GH receptors (4, 5) to induce differentiation of an immature ductal tree into more mature terminal end buds (TEBs) and alveolar structures. TEBs extend into the mammary fat pad and lead to further ductal morphogenesis. This process, which requires the synergy of GH and E2, probably involves local production of insulin-like growth factor I (IGF-I) that mediates the action of GH (6, 7). Stromal elements must be present for mammary epithelial elements to mature (8 –11). Based on the facts that IGF-I can substitute for GH in mammary development in hypophysectomized rats, that GH induces IGF-I messenger RNA (mRNA) in mammary gland, and that GH causes differentiation of adipose cells (12–15), we have hypothesized that stromal elements within the mammary gland mediate at least some of the actions of GH in pubertal mammary development by stimulating IGF-I mRNA and the IGF-I protein within the stroma, which, in turn, act on glandular elements in a paracrine fashion. To test the part of this hypothesis that addresses the site of action of GH, we determined effects of GH on IGF-I mRNA production in three types of tissue: 1) gland-free mammary stromal tissue, 2) whole mammary glands from which the gland-free fat pads were removed, and 3) subscapular fat pads. The results are reported here. Materials and Methods Animals Female Sprague-Dawley rats were hypophysectomized and oophorectomized at 21 days of age, as previously described (7). At 45 days of age, groups of 10 animals received either a single injection of bovine GH (bGH; 100 mg) or a saline control injection ip. Some animals were also given E2 in SILASTIC brand capsules (Dow Corning, Midland, MI) implanted sc (16). After 12 h, the point at which maximal stimulation of IGF-I mRNA occurs (4), animals were killed, and both lumbar mammary glands were removed. The mammary glands were divided into the gland-free mammary fat pad (Fig. 1) and the remainder of the mammary gland containing the glandular epithelial elements. Figure 2 is a photomicrograph of a lumbar mammary gland with the mammary fat pad separated from the remainder of the gland. The subscapular fat pad was also removed. mRNA isolation Immediately after removal from the animals, the mammary gland and subscapular fat pad tissue were snap-frozen in liquid nitrogen. Total RNA was prepared by the acid guanidine phenol chloroform extraction method (17). RNA was used for solution hybridization and/or RT-PCR experiments. Received August 29, 1997. Address all correspondence and requests for reprints to: David L. Kleinberg, M.D., Room 16043W, DVA Medical Center, 423 East 23rd Street, New York, New York 10010. E-mail: . nyu.edu. 659 660 GH ACTION ON FAT PAD IN MAMMARY DEVELOPMENT FIG. 1. Schematic of rat with lumbar mammary glands divided into the gland-free and gland-rich areas and the approximate location of the subscapular fat pad. Endo • 1998 Vol 139 • No 2 bGH plus E2, and E2 alone compared to control values in gland-free mammary fat pads and whole mammary glands. GH significantly stimulated IGF-I mRNA, and E2 enhanced that effect. That the effect was equal in gland-free mammary fat pads suggests that stromal tissue may mediate the effect of GH in the mammary gland. Figure 4 compares effects of the above combinations of hormones on IGF-I mRNA production in subscapular fat pads and gland-free mammary fat pads. In the former, bGH had a more pronounced effect on IGF-I mRNA then in the latter, and there was no synergism with E2. The quantitative nature of the RT-PCR assay was confirmed by solution hybridization/ribonuclease protection. A representative solution hybridization gel shown in Fig. 5 validates the stimulatory effect of bGH on IGF-I mRNA production in whole mammary gland, gland-free mammary fat pads, and subscapular fat pads. The solution hybridization data also confirm the synergistic effects of E2 in whole mammary gland and mammary fat pads and the nonsynergistic effects of E2 in subscapular fat pads. Discussion Solution hybridization The full-length rat IGF-I complementary DNA (cDNA) fragment (18) was isolated by EcoRI digestion and subcloned into the EcoRI site of the vector pcDNA3 (Invitrogen, San Diego, CA) in the sense orientation with respect to the cytomegalovirus promoter. The resulting pcDNA3-IGF-I construct was used to synthesize sense and antisen (...truncated)